We propose to study the pathophysiology of focal and global cerebral ischemia using the respective clincal models of (1) ischemic stroke due to occlusive cerebrovascular disease and (2) cardiac arrest followed by resuscitation and to compare and contrast these two states. 31P NMR spectroscopy will be employed to measure brain energy status, pH, Mg2+ and phospholipids. Proton spectorscopy will be employed to measure brain lactate. Infarct volume and brain edema will be assessed using CT and MR imaging techniques. CBF will be measured using inhalation or IV Xenon, SPECT or NMR techniques. All measures will be brought to bear on the assessment of brain acidosis and its role in ischemic cell damage and influence on neurological outcome. The issue of high systemic glucose levels and its potential influence in worsening brain acidosis, ishcemic cell damage and neurological outcome is also addressed. The mechanisms and meaning of the "flip-flop" from acidosis to alkalosis that occurs during the progression of ischemia will be explored particularly in relation to reperfusion and brain edema. The documentation of this pH "flip-flop" will be used as a marker of a therapeutic window of time during which control of blood glucose might favorably effect metabolic and neurological outcome. Brain buffering capacity to a paCO2 challenge will be assessed in patients with brain alkalosis following global cerbral ischemia to evaluate if active cellular buffering mechanisms can be distinguished from metabolic paralysis.